AWS's managed MCP server: call any AWS API, run sandboxed Python against your account, and load AWS agent skills on demand — under IAM, CloudWatch and CloudTrail.
NetBox MCP Server
Summary
NetBox Labs' official read-only MCP server — query your network source of truth by object type and filter, and read the change audit trail from an agent.
Features
- Query core NetBox objects by type with filters
- Fetch a single object's full record by ID
- Read the change audit trail with filtering
- Auto-discovery of plugin object types
- Read-only by default — no write surface to misuse
- stdio and HTTP transports
Installation
Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.
Description
NetBox is the network source of truth that most infrastructure teams end up building their automation around, and this official server from NetBox Labs makes it readable by an agent without giving it a way to break anything.
The tool surface is deliberately three tools wide. get_objects retrieves core NetBox objects by type with filters; get_object_by_id pulls the full record for one object; get_changelogs reads the audit trail of who changed what and when. That is enough to answer the questions people actually ask a source of truth — which devices sit in this rack, what is on this VLAN, which prefixes are free, when did this interface description change — while GraphQL, dynamic model discovery and write access stay explicitly out of scope. Plugin object types can be auto-discovered on top of the always-available core types.
The project states its own design goal as "easy to get started with, hard to misuse": read-only by default, no plugin surface, and Apache 2.0 licensed so that forking is a first-class path for anyone who needs more.
Run it over stdio or HTTP with NETBOX_URL and a read-only NETBOX_TOKEN. Note the v1.0.0 breaking change: the entry point is now netbox-mcp-server, not server.py.
Related MCP Servers
Google's official MCP server for driving the gcloud CLI from natural language, with allowlist and denylist controls over which commands an agent may run.
Google's open-source MCP server for enterprise databases — one binary that fronts AlloyDB, BigQuery, Spanner, Postgres, MySQL, Oracle, MongoDB, Snowflake and more.
Datadog's official managed MCP server — query logs, metrics, traces, incidents and dashboards from an AI agent, inside your existing Datadog permissions.
